arXiv subjects

Dian Shen

Publications and source records attributed to Dian Shen.

7 recordsLinked to original sources

arXiv2026

QuReC: All-in-One Image Restoration with Query-Specific Guidance and Local-Global Response Calibration

All-in-one image restoration aims to recover clean images degraded by multiple corruption types using a single unified model. Existing methods typically rely on image-level prompts or shared guidance to handle diverse degradations. However, such a paradigm becomes inadequate when degradations are spatially heterogeneous or even coexist in mixed forms within a single image. Yet spatially adaptive guidance alone is not sufficient, since accurate restoration also requires each spatial query to reliably aggregate complementary information from local neighborhoods and global contexts. To this end, we propose QuReC, a unified framework for all-in-one image restoration. QuReC consists of a Degradation-Guided Query Reconstruction Module (DQRM) and a Local-Global Response Calibration Module (LGRCM). Specifically, DQRM matches each spatial query against a degradation prototype space to reconstruct a query-specific degradation-aware representation, thereby providing fine-grained spatially adaptive restoration guidance. To further stabilize this query-wise matching process, we introduce a weakly supervised prototype matching learning strategy to improve optimization stability and degradation semantic consistency. Meanwhile, LGRCM performs local-global dual-branch aggregation and calibrates the aggregated responses with learnable priors, improving the reliability of feature aggregation and the coordination between local detail modeling and global context modeling. Extensive experiments demonstrate that QuReC achieves superior performance on multiple all-in-one image restoration benchmarks. The code is released at https://github.com/zhoushen1/QuReC.

cs.CV↗

arXiv2026

ToolCaching: Towards Efficient Caching for LLM Tool-calling

Recent advances in Large Language Models (LLMs) have revolutionized web applications, enabling intelligent search, recommendation, and assistant services with natural language interfaces. Tool-calling extends LLMs with the ability to interact with external APIs, greatly enhancing their practical utility. While prior research has improved tool-calling performance by adopting traditional computer systems techniques, such as parallel and asynchronous execution, the challenge of redundant or repeated tool-calling requests remains largely unaddressed. Caching is a classic solution to this problem, but applying it to LLM tool-calling introduces new difficulties due to heterogeneous request semantics, dynamic workloads, and varying freshness requirements, which render conventional cache policies ineffective. To address these issues, we propose ToolCaching, an efficient feature-driven and adaptive caching framework for LLM tool-calling systems. ToolCaching systematically integrates semantic and system-level features to evaluate request cacheability and estimate caching value. At its core, the VAAC algorithm integrates bandit-based admission with value-driven, multi-factor eviction, jointly accounting for request frequency, recency, and caching value. Extensive experiments on synthetic and public tool-calling workloads demonstrate that ToolCaching with VAAC achieves up to 11% higher cache hit ratios and 34% lower latency compared to standard policies, effectively accelerating LLM tool-calling in practical applications.

cs.SE↗

arXiv2025

AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices

On-device machine learning (ML) has become a fundamental component of emerging mobile applications. Adaptive model deployment delivers efficient inference for heterogeneous device capabilities and performance requirements through customizing neural architectures. SuperNet-based approaches offer a promising solution by generating a large number of model variants from a pre-trained ML model. However, applying SuperNet in existing frameworks suffers from tedious model-aware development and time-consuming hardware-aware profiling, which limits their practical adoption. We present AutoTailor, the first framework to enable automated, end-to-end SuperNet-based adaptive model deployment for edge devices. Unlike manual SuperNet construction, AutoTailor employs a computation graph-guided compilation approach to automatically transform user-provided ML models into SuperNets. To support efficient specialization, AutoTailor incorporates learning-free latency and accuracy predictors, enabling low-cost yet accurate performance prediction. Our extended evaluations demonstrate that AutoTailor reduces the lines of code for SuperNet construction by 11--27$\times$, decreases hardware-aware profiling costs by at least 11$\times$, and achieves up to 15.60\% absolute accuracy improvement and 60.03\% latency reduction compared to state-of-the-art approaches across diverse models and devices.

cs.LG↗

arXiv2025

Joint Optimization of DNN Model Caching and Request Routing in Mobile Edge Computing

Mobile edge computing (MEC) can pre-cache deep neural networks (DNNs) near end-users, providing low-latency services and improving users' quality of experience (QoE). However, caching all DNN models at edge servers with limited capacity is difficult, and the impact of model loading time on QoE remains underexplored. Hence, we introduce dynamic DNNs in edge scenarios, disassembling a complete DNN model into interrelated submodels for more fine-grained and flexible model caching and request routing solutions. This raises the pressing issue of jointly deciding request routing and submodel caching for dynamic DNNs to balance model inference precision and loading latency for QoE optimization. In this paper, we study the joint dynamic model caching and request routing problem in MEC networks, aiming to maximize user request inference precision under constraints of server resources, latency, and model loading time. To tackle this problem, we propose CoCaR, an offline algorithm based on linear programming and random rounding that leverages dynamic DNNs to optimize caching and routing schemes, achieving near-optimal performance. Furthermore, we develop an online variant of CoCaR, named CoCaR-OL, enabling effective adaptation to dynamic and unpredictable online request patterns. The simulation results demonstrate that the proposed CoCaR improves the average inference precision of user requests by 46% compared to state-of-the-art baselines. In addition, in online scenarios, CoCaR-OL achieves an improvement of no less than 32.3% in user QoE over competitive baselines.

cs.NI↗

arXiv2025

Unlocking Full Efficiency of Token Filtering in Large Language Model Training

Token filtering has been proposed to enhance the utility of large language models (LLMs) by eliminating inconsequential tokens during training. While usingfewer tokens is expected to reduce computational workloads, existing methods have not yet achieved a real-world efficiency boost. This is primarily due to two factors: (1) existing work has inadequate sparsity for speedup, and (2) token filtering operates within a sparsity range that is non-standard in existing machine learning (ML) libraries and thus cannot be efficiently supported. This paper presents Centrifuge, a system that leverages algorithm and system co-design to unleash the full efficiency of token filtering in LLM training. At the algorithm level, Centrifuge filters activations of inconsequential tokens in the attention backward kernel to amplify the sparsity in backward computation. At the system level, Centrifuge proposes an automatic workflow that transforms sparse GEMM into dimension-reduced dense GEMM for optimized efficiency using standard ML libraries. Evaluations on models with various scales--from 1.1B to 40B--demonstrate that Centrifuge reduces backpropagation time by up to 49.9\% and end-to-end training time by up to 34.7\% when filtering 50\% of tokens. Utility assessments indicate that Centrifuge preserves the utility benefits of token filtering and significantly enhances model performance by up to 26.6\% compared to standard training. Centrifuge is designed for seamless integration into existing LLM training frameworks, enabling systems already utilizing token filtering to accelerate training with just one line of code.

cs.LG↗

arXiv2025

Swift: Rethinking RDMA Control Plane for Elastic Computing

Elastic computing enables dynamic scaling to meet workload demands, and Remote Direct Memory Access (RDMA) enhances this by providing high-throughput, low-latency network communication. However, integrating RDMA into elastic computing remains a challenge, particularly in control plane operations for RDMA connection setup. This paper revisits the assumptions of prior work on high-performance RDMA for elastic computing, and reveals that extreme microsecond-level control plane optimizations are often unnecessary. By challenging the conventional beliefs on the slowness of user-space RDMA control plane and the difficulty of user-space RDMA resource sharing, we uncover new design opportunities. Our key insight is that user-space RDMA connection setup can be significantly improved with caching, while RDMA resources can be efficiently shared among processes using fork. In light of this, we propose Swift, a simple yet effective solution that co-designs RDMA with a serverless framework to optimize performance for elastic computing. At its very core, Swift handles cold and warm serverless requests by swiftly initializing the RDMA control plane with cache-optimized libibverbs, and manages fork requests by leveraging the RDMA's fork capability. Implemented with OpenWhisk, Swift delivers 30.56-46.50% higher average throughput and 18.55-37.21% lower latency, at a cost of 6.5% control plane overhead, compared to prior solutions.

cs.NI↗

arXiv2025

FLASH-FHE: A Heterogeneous Architecture for Fully Homomorphic Encryption Acceleration

While many hardware accelerators have recently been proposed to address the inefficiency problem of fully homomorphic encryption (FHE) schemes, none of them is able to deliver optimal performance when facing real-world FHE workloads consisting of a mixture of shallow and deep computations, due primarily to their homogeneous design principle. This paper presents FLASH-FHE, the first FHE accelerator with a heterogeneous architecture for mixed workloads. At its heart, FLASH-FHE designs two types of computation clusters, ie, bootstrappable and swift, to optimize for deep and shallow workloads respectively in terms of cryptographic parameters and hardware pipelines. We organize one bootstrappable and two swift clusters into one cluster affiliation, and present a scheduling scheme that provides sufficient acceleration for deep FHE workloads by utilizing all the affiliations, while improving parallelism for shallow FHE workloads by assigning one shallow workload per affiliation and dynamically decomposing the bootstrappable cluster into multiple swift pipelines to accelerate the assigned workload. We further show that these two types of clusters can share valuable on-chip memory, improving performance without significant resource consumption. We implement FLASH-FHE with RTL and synthesize it using both 7nm and 14/12nm technology nodes, and our experiment results demonstrate that FLASH-FHE achieves an average performance improvement of $1.4\times$ and $11.2\times$ compared to state-of-the-art FHE accelerators CraterLake and F1 for deep workloads, while delivering up to $8.0\times$ speedup for shallow workloads due to its heterogeneous architecture.

cs.AR↗